Databox for AI agents
Databox pulls data from many tools into shared dashboards, letting a team track KPIs and trends in one place instead of switching between apps. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in Databox.
Push data into a dataset
Push a new data point into a dataset to keep a dashboard's metric current.
Check a dataset's ingestion status
Check whether recently pushed data has finished processing into a shared dashboard.
Confirm before deleting a data source
Confirm before deleting a data source, since every dashboard using it stops updating.
7 Databox actions, graded by risk.
Every Databox action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
2Look things up. Allowed by default.
- databox.list_accountsList Accounts
- databox.get_dataset_ingestion_statusGet Dataset Ingestion Status
write
3Create and change things. Allow, or ask the user first.
- databox.create_datasetCreate Dataset
- databox.create_data_sourceCreate Data Source
- databox.push_data_v1Push Data to Dataset (V1)
destructive
2Delete, cancel or archive. Ask first, or deny outright.
- databox.delete_datasetDelete Dataset
- databox.delete_data_sourceDelete Data Source
Databox in three steps.
- 01Your users connect DataboxThey add their Databox api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create Dataset” can wait for the user, and “delete Dataset” can be denied outright.
- 03Any agent can actYour agent calls Databox through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('databox', { read: 'allow', write: 'ask', // create_dataset destructive: 'deny', // delete_dataset }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Databox connects.
Users add their Databox api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same Databox connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Databox from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
Databox and Arc0, answered.
Can I use Databox with Claude, ChatGPT or Cursor?
Yes. Connect Databox to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Databox actions you allow.
How do users connect Databox?
Users add their Databox api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which Databox actions can my agent take?
7 in total: 2 read, 3 write and 2 destructive, such as “create Dataset”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Databox?
Yes. Actions like “delete Dataset” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.
Can my own backend call Databox too?
Yes. The same Databox connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Databox into your agent.
Your users connect Databox once, under your brand. Your agent gets 7 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan